arXiv (All)
open access
Who Sets the Pace of Judge Hacking? Judge Capability versus the Policy-Judge Gap
When a policy is trained with GRPO against a frozen, reference-free LLM judge, the judge accepts more of its answers, but in a post-hoc analysis we detect no gain in held-out greedy accuracy: in all 12 (policy, judge) cells, the seed-mean gain never exceeded 0.04. A gold-reward control with the same
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- Court Review: Journal of the American Judges Association, Vol. 61, No. 2#420094
- 28 U.S.C. § 2505 — Trial before judges#556718
- Deepfakes in Court: How Judges Can Proactively Manage Alleged AI-Generated Material in National Security Cases#451751
- The First Token Is Not the Verdict: Hidden Costs of Reading LLM Judges Without Generating#1154677
- JEV vs. LLMs as Rubric Judges: Cheaper, Faster, and Wrong in the Same Places#1077696
Record · ID 1327028
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive